Your backup is not a backup until you've restored from it

Picture a dental lab on a Tuesday morning. Twelve crowns due Friday, every one of them a CAD file on the workstation in the corner, and the workstation won’t turn on.

The owner stays calm, because there’s a backup. There has been a backup for six years; it costs $40 a month and it has never once been needed. Today it’s needed, and today is when everyone learns the backup stopped including the CAD folder after a software update moved it, four years ago.

That story repeats in some version every week somewhere, a practice, a law office, a shop. The backup existed. The protection didn’t.

A backup is a claim. A restore is proof.

Backup software is good at reporting success. It ran on schedule, it copied what it was told, the dashboard is green. What the dashboard can’t tell you:

• whether the folders being copied are the ones your business actually lives in
• whether the files come back uncorrupted and openable
• whether anyone in the building knows the password to the backup account
• how long a full recovery takes, an hour or a week

The only way to learn any of that is to do a restore before you have to. Pick real files, bring them back from the backup onto a different machine, and open them. If a crown design opens in the CAD software, that file was protected. Everything else is assumption.

What a restore test looks like

A real test takes under an hour for a small office:

1. Choose targets that matter. Not a random text file, the patient database, this month’s billing folder, the design files.
2. Restore to a different location. You’re proving the backup works, not gambling the originals on it.
3. Open everything you restored. Files that exist but won’t open count as failures.
4. Time it and write it down. The point of the note is the next test: if restores suddenly get slower or smaller, something changed upstream.

How often is enough?

Quarterly, as a floor. Data moves, software updates relocate folders, and the person who configured the backup changes jobs. A quarterly restore catches the quiet failures while they’re still trivia instead of catastrophe.

If a quarter sounds like a lot, weigh it against the other side of the scale: the going recovery cost when there is no working backup is measured in weeks of rebuilding, and some of what’s lost, patient history, years of designs, doesn’t rebuild at any price.

One more thing the test catches: ransomware math

Modern ransomware doesn’t just lock the files on the machine it lands on; it reaches for every drive and synced folder it can see. A cloud-sync drive faithfully syncs the encrypted versions, sync is not backup. What beats ransomware is versioned history stored where the malware can’t write, so you can step back to the day before. A restore test is also how you confirm your history actually reaches back far enough.

What to ask your IT person this week

You don’t need to be technical, three questions do it:

1. “When did we last restore real files from our backup, and did they open?”
2. “How far back does our version history go?”
3. “If this office flooded tonight, what’s running at 8 AM, and on what?”

Confident, specific answers mean you’re in good hands. A pause means it’s time for a second opinion.

We run quarterly restore tests as a standing part of backup and recovery at Saibeo, and we show clients the results. If you’re in the Oklahoma City area and the three questions above made your stomach drop, book a free assessment, thirty minutes, and you’ll know exactly where you stand.